summaryrefslogtreecommitdiff
path: root/sys/netinet/tcp_syncache.c
Commit message (Expand)AuthorAgeFilesLines
* MFC r211327:Andre Oppermann2010-08-221-5/+24
* MFC r210666:Andre Oppermann2010-08-141-1/+2
* MFC r207369:Bjoern A. Zeeb2010-05-061-12/+7
* MFC r204143:Bjoern A. Zeeb2010-03-271-2/+24
* Merge the remainder of kern_vimage.c and vimage.h into vnet.c andRobert Watson2009-08-011-1/+0
* Somewhere along the line accept sockets stopped honoring theJulian Elischer2009-07-281-4/+2
* Remove unused VNET_SET() and related macros; only VNET_GET() isRobert Watson2009-07-161-3/+3
* Build on Jeff Roberson's linker-set based dynamic per-CPU allocatorRobert Watson2009-07-141-49/+31
* Replace struct tcpopt with a proxy toeopt struct in the TOE driver interface toLawrence Stewart2009-07-131-4/+17
* Introduce an infrastructure for dismantling vnet instances.Marko Zec2009-06-081-0/+13
* Move "options MAC" from opt_mac.h to opt_global.h, as it's now in GENERICRobert Watson2009-06-051-1/+0
* Add internal 'mac_policy_count' counter to the MAC Framework, which is aRobert Watson2009-06-021-2/+0
* Permit buiding kernels with options VIMAGE, restricted to only a singleMarko Zec2009-04-301-0/+3
* Update stats in struct tcpstat using two new macros, TCPSTAT_ADD() andRobert Watson2009-04-111-24/+24
* Since we allow conditional allocation of labels on syncache entries,Robert Watson2009-01-111-2/+0
* Use inc_flags instead of the inc_isipv6 alias which so farBjoern A. Zeeb2008-12-171-15/+16
* Another step assimilating IPv[46] PCB code - directly useBjoern A. Zeeb2008-12-151-4/+4
* Rather than using hidden includes (with cicular dependencies),Bjoern A. Zeeb2008-12-021-0/+1
* Merge more of currently non-functional (i.e. resolving toMarko Zec2008-11-261-9/+13
* Change the initialization methodology for global variables scheduledMarko Zec2008-11-191-5/+11
* Remove endearing but syntactically unnecessary "return;" statementsRobert Watson2008-10-261-2/+0
* Fix a number of style issues in the MALLOC / FREE commit. I've tried toDag-Erling Smørgrav2008-10-231-2/+2
* Retire the MALLOC and FREE macros. They are an abomination unto style(9).Dag-Erling Smørgrav2008-10-231-2/+1
* Step 1.5 of importing the network stack virtualization infrastructureMarko Zec2008-10-021-12/+37
* Implement IPv6 support for TCP MD5 Signature Option (RFC 2385)Bjoern A. Zeeb2008-09-131-1/+1
* Move some struct defs around. This is a prep step for Vimage.AJulian Elischer2008-08-251-66/+0
* Cache the cred locally in _syncache_add() while holding the locks, soBjoern A. Zeeb2008-08-231-0/+12
* Add an explicit comment why we NULLify the two variables.Bjoern A. Zeeb2008-08-231-0/+1
* A bunch of formatting fixes brough to light by, or created by the Vimage commitJulian Elischer2008-08-201-2/+4
* Commit step 1 of the vimage project, (network stack)Bjoern A. Zeeb2008-08-171-78/+79
* Minor style tweaks.John Baldwin2008-08-051-6/+5
* MFp4 (//depot/projects/tcpecn/):Rui Paulo2008-07-311-1/+12
* add interface for external consumers to syncache_expand - rename syncache_add...Kip Macy2008-07-211-1/+14
* Change incorrect stale cookie detection in syncookie_lookup() that prematurelyStephan Uphoff2008-06-161-1/+1
* Fix a check in SYN cache expansion (syncache_expand()) to accept packets that...Stephan Uphoff2008-06-161-4/+7
* Add code to allow the system to handle multiple routing tables.Julian Elischer2008-05-091-0/+4
* Always bump tcpstat.tcps_badrst if we get a RST for a connection in theJohn Baldwin2008-05-081-4/+5
* move tcbinfo lock acquisition in to syncacheKip Macy2008-04-191-0/+3
* Convert pcbinfo and inpcb mutexes to rwlocks, and modify macros toRobert Watson2008-04-171-6/+6
* Fix bugs in the TCP syncache timeout code. including:Ruslan Ermilov2007-12-191-8/+18
* incorporate feedback since initial commitKip Macy2007-12-171-14/+22
* Add interface for tcp offload to syncache:Kip Macy2007-12-121-15/+47
* Remove spurious timestamp check. RFC 1323 explicitly states that timestamps MAYKip Macy2007-12-121-17/+1
* Add padding for anticipated functionalityKip Macy2007-12-071-0/+3
* Fix SACK negotiation that was broken in rev 1.105.Mike Silbersack2007-12-041-1/+1
* Move call to tcp_signature_compute() after we adjusted the payload offsetBjoern A. Zeeb2007-11-301-5/+6
* Comment out the syncache's test which ensures that hosts which negotiate TCPMike Silbersack2007-11-201-0/+6
* Normalize TCP syncache-related MAC Framework entry points to match mostRobert Watson2007-10-251-6/+6
* Merge first in a series of TrustedBSD MAC Framework KPI changesRobert Watson2007-10-241-1/+1
* Pick the smallest possible TCP window scaling factor that will still allowMike Silbersack2007-10-191-6/+17